Cloud-Based Dataset Analysis for the Social Sciences

📄 Project Abstract

This project investigates the applicability of Cloud Computing to the analysis of large datasets in the growing area of e-Social Science. Social scientists are procuring data from increasingly diverse sources, allowing new social simulations of populations on a scale not seen before. These simulations are demanding increasing amounts of computational resources, for which grids have to date been the preferred solution. Public clouds have emerged offering a pay-by-use service supplying compute and storage resources, charged for in quantitative chunks. The potential for highly parallelised processing of data using these virtualised resources is examined in the context of the social sciences and an attempt is made at producing a set of requirements that developers of applications in this research area should consider.